I think that's true - but my experience was that it was a good language for beginners, but didn't "scale" well as you started working on more complex programs. The English-like syntax could get in the way as soon as you started to assemble more complex expressions, and you'd be left hunting for the right syntax as if you were playing an old Infocom game. It gave the impression that the parser was more accommodating and sophisticated than it really was, when really it had a rigid syntax like most other programming languages. And because most of the documentation was aimed at beginners, it was hard to find clear answers about _exactly_ what the syntax had to look like when assembling those more complex expressions. It has been a long time since I was writing HyperTalk, but I remember being very frustrated that I couldn't find a formal language spec (in BNF or a bubble diagram or something) the way I could for languages like Pascal.
Still, it really should get a lot of credit for encouraging a lot of beginners to write code and build quite sophisticated little databases complete with all kinds of animations and visual effects that would have been monstrously difficult to code from scratch in C or Pascal. I should also remember that as a programmer and Computer Science student, _I was not the real target audience_ for the language design.